The design of enzymes with tailored physical and catalytic properties is one of the fundamental thrusts of modern protein science, with the potential for profound technological and medical impact. Laboratory directed evolutionary approaches along with rational design principles have now been successfully applied to enhance protein properties and function. What remains is the important next step of rolling up our sleeves and attacking important problems with an eye toward details, which are likely to be enzyme specific. This proposal extends our enzyme directed evolution program in important enabling directions in the area of engineered proteases. Our most recent work with the OmpT protease represents by far the most general manipulation of P1 and P1' substrate specificity of a protease while retaining high overall levels of catalytic activity. Beyond their use in the detergent industry, engineered proteases have tremendous practical potential as either proteomic tools or catalytic therapeutics. In particular, proteases specific for substrates containing modifications such as phosphorylation or O- GlcNAc would represent useful new tools for identifying modified proteins in high throughput proteomics assays. Under Specific Aim 1, we will engineer OmpT variants specific for cleaving only substrates containing phosphorylated or O-GlcNAc serine. We will also investigate whether we can engineer restriction-like proteases that can very selectively recognize an extended sequence comprising residues well beyond P1 and P1'. Under Specific Aim 2, we will extend precise OmpT protease recognition to include P2, P3, P2', and P3'. In particular, we will target Gln-His-Ala-Arg-Ala-Ser (QHA$RAS), residues 68-73 of the C-terminus of the C3a anaphylatoxin peptide. C- terminal cleavage of C3a interferes with its biological effects in complement activation. We recognize that creating highly specific subsites in OmpT by mutagenesis and sorting is an exciting yet risky goal and that an engineered OmpT is an unlikely therapeutic clinical candidate because of its bacterial origin. Therefore, for Specific Aim 3, we will engineer precise C3a cleavage activity into the secreted human trypsin-like protease granzyme A in hopes of producing a clinical candidate. As a practical deliverable, following the functional assays of our best variants (Section D6) we will, for the first time, be able to validate the proteolytic approach to complement inhibition, applicable to a wide variety of inflammatory disease therapies.